The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted data from the local environment to generate commit summaries.
Ingestion points: The skill reads output from git diff --staged and git diff (SKILL.md).
Boundary markers: The instructions lack explicit delimiters to separate the diff output from the agent's internal instructions, nor do they instruct the model to ignore embedded instructions within the code changes.
Capability inventory: The skill has the capability to read local files via git diff and modify the repository using git commit (SKILL.md).
Sanitization: No sanitization or filtering of the git diff output is performed before it is processed by the agent.
[COMMAND_EXECUTION]: The skill uses standard local shell commands to analyze the repository state and finalize commits.
Evidence: The workflow in SKILL.md explicitly directs the agent to execute git diff to understand modifications and git commit to apply the formatted message.

commit-helper